Abstract: A method and system for providing end-to-end multi-task denoising for joint signal distortion ratio (SDR) and perceptual evaluation of speech quality (PESQ) optimization is herein disclosed. According to one embodiment, an method includes receiving a noisy signal, generating a denoised output signal, determining a signal distortion ratio (SDR) loss function based on the denoised output signal, determining a perceptual evaluation of speech quality (PESQ) loss function based on the denoised output signal, and optimizing an overall loss function based on the PESQ loss function and the SDR loss function.

Abstract: An electronic apparatus and a method thereof are provided. The electronic apparatus according to an embodiment includes a memory to store a speaker model including characteristic information of a voice of a user corresponding to a specific word, and a processor configured to, based on a first input voice corresponding to the specific word, provide a voice recognition service through the electronic apparatus based on a first authentication performed based on data on the first input voice and the speaker model, and perform a second authentication based on data corresponding to the specific word among data of the second input voice that is input while the voice recognition service is provided and the speaker model.

Abstract: Systems, methods, and devices for privacy protection and user data obfuscation are disclosed. A speech-controlled device captures audio including a spoken command, and sends audio data corresponding thereto to a server(s). The server(s) 120 determines a user that spoke the command. The server(s) also determines, based on a profile of the user, user data (e.g., age, geographic location, etc.). The server(s) determines user group data encompassing the user data (e.g., including an age range encompassing the user's age, a geographic area encompassing the user geographic location, etc.). The server(s) determines a remote device(s) storing or having access to content responsive to the spoken command. The server(s) sends the user group data to the remote device(s), receives output content responsive to the spoken command and tailored to the user group data from the remote device(s), and causes the speech-controlled device to emit the output content.

Abstract: Computer generated speech can be generated for cross-lingual natural language textual data streams by utilizing a universal phoneme set. In a variety of implementations, the natural language textual data stream includes a primary language portion in a primary language and a secondary language portion that is not in the primary language. Phonemes corresponding to the secondary language portion can be determined from a set of phonemes in a universal data set. These phonemes can be mapped back to a set of phonemes for the primary language. Audio data can be generated for these phonemes to pronounce the secondary language portion of the natural language textual data stream utilizing phonemes associated with the primary language.

Abstract: Disclosed is an approach for allowing an entity to perform semantic analysis upon private data possessed by an enterprise, and to automatically perform categorization of that data for processing within the enterprise. A semantic API can be provided to allow the enterprise to provide the private data to a sematic analysis system, even when the semantic analysis system is configured as a multi-tenant system that handles other items of public or private data. A rules-based routing architecture may be provided to facilitate analysis and routing of analyzed messages to the appropriate destination within the organization.

Abstract: Systems and methods for e-commerce systems using natural language understanding are described. A computing device is configured receive a user utterance including at least one identified semantic component and at least one missing semantic component and generate a context stack including a set of context entries. Each of the context entries includes a root intent element, an entity list element, and a dialogue stack and each context entry in the set of context entries is associated with one of a user utterance or a system utterance. The computing device is further configured to review at least one context entry in the set of context entries to locate the at least one missing semantic element within the dialogue stack and generate an intent flow execution request including the at least one semantic element from the first speech data and the missing semantic element.

Abstract: According to an exemplary embodiment of the present disclosure, a method, apparatus for evaluating a review, a device and a computer readable storage medium are provided. The method for evaluating a review includes obtaining a first vectorized representation of a set of text items in a review for a target object. The method further includes extracting a semantic feature of at least two consecutive text items of the set of text items based on the first vectorized representation. The method further includes determining a degree of importance of the at least two consecutive text items in a context of the review, and determining a degree of the review helping a user to evaluate the target object, based on the degree of importance and the semantic feature. In this way, an automatic, efficient, and accurate evaluation of the value of a review may be achieved.

Abstract: A token is extracted from a Natural Language input. A phonetic pattern is computed corresponding to the token, the phonetic pattern including a sound pattern that represents a part of the token when the token is spoken. New data is created from data of the phonetic pattern, the new data including a syllable sequence corresponding to the phonetic pattern. A state of a data storage device is changed by storing the new data in a matrix of syllable sequences corresponding to the token. An option is selected that corresponds to the token by executing a fuzzy matching algorithm using a processor and a memory, the selecting of the option is based on a syllable sequence in the matrix.

Abstract: There are provided a device and method for topic modeling with sentiment priors based on distributed representations. A method for topic modeling with sentiment priors based on distributed representations includes: inputting a review into a topic model; and by the topic model, determining a representation for each word in the review, wherein the representations are word vectors in a semantic space; and estimating the representations using the sentiment priors to determine a topic corresponding to the review, wherein the topic model includes the sentiment priors trained on the basis of the representations distributed by using a regularizer, the regularizer defining the same sentiment to words having similar word vectors, and wherein each sentiment prior is the same for words having similar word vectors.

Abstract: Expansion of intent classification data utilizing batch utterance scheduling, by a processor in a computing environment. A set of unlabeled examples for intent processing is received by an intent builder iteratively defining an intent. The set of examples are separated into a first subset processed according to a first model and a second subset processed according to a second model. The first subset is incorporated into the intent builder during a building iteration and scheduling a first batch processing of the second subset processed according to the second model based on a scheduling criteria. The first batch processing of the second subset is initiated once the scheduling criteria is satisfied. Upon completion of the first batch processing, results of the completion are used to influence additional examples retrieved from the first subset and the second subset during a subsequent building iteration by the intent builder.

Abstract: Apparatus for post-processing an audio signal, including: a converter for converting the audio signal into a time-frequency representation; a transient location estimator for estimating a location in time of a transient portion using the audio signal or the time-frequency representation; and a signal manipulator for manipulating the time-frequency representation, wherein the signal manipulator is configured to reduce or eliminate a pre-echo in the time-frequency representation at a location in time before the transient location or to perform a shaping of the time-frequency representation at the transient location to amplify an attack of the transient portion.

Abstract: Systems and processes for accelerating task performance are provided. An example method includes, at an electronic device with a display, displaying a candidate shortcut affordance associated with a user activity, detecting a first set of inputs corresponding to a selection of the candidate shortcut affordance, in response to detecting the first set of inputs, displaying a first set of candidate task affordances, detecting a second set of inputs corresponding to a selection of a candidate task affordance associated with a first task, displaying a second set of candidate task affordances, detecting a third set of inputs corresponding to a selection of a candidate task affordance associated with a second task, and in response to detecting the second set of inputs and the third set of inputs, associating the first task and the second task with a task sequence for a voice shortcut corresponding to the user activity.

Abstract: A system for state dependency-based task execution and natural language response generation is disclosed. The system includes an input request acquisition subsystem to obtain a plurality of natural language input requests from one or more sources through at least one information capturing medium; a state definition subsystem to determine a plurality of states for a plurality of obtained natural language input requests; a state dependency definition subsystem to create a list of a plurality of determined states required to generate a dependency order, to analyse a generated dependency order of the plurality of determined states, to generate a dynamic working procedure to generate a temporary natural language response; a state based response generation subsystem to generate a structured natural language response dynamically upon successful execution of the dynamic working procedure.

Abstract: Chat discourse convolution can include performing a lexicographic analysis of the content of a chat message of a chat discourse and storing a result of the lexicographic analysis in a database. The convolution can further include generating a hash of the content of the chat message and generating a leximarker. The leximarker can encapsulate a reference to the stored lexicographic analysis as well as the hash of the content of the chat message.

Abstract: Exemplary embodiments automate the production of memos in support of the due diligence process for underwriting. Target users may include investment bankers, financial analysts, and the like. The data collection process of supporting data for the memo may be automated thus improving efficiency in the creation of the due diligence memo. Both the collected data and memos may be stored in a database, enabling access at a later date and enabling the collected data from one memo, as well as other information from the memo, to be used for future memos. The data collection may be conducted from both structured and unstructured sources. The sources may be both internal and external sources. The external sources may be public data sources.

Abstract: An information processing apparatus includes: a receiver configured to receive an utterance content of a speaker, a processing structure of a work in which the speaker utters, the work including plural processing units, and a processing unit in execution in the processing structure; an extraction unit configured to extract a related document including a sentence whose similarity to the utterance content of the speaker received by the receiver is equal to or higher than a threshold, from among related documents that are associated in advance with at least one processing unit including the processing unit in execution received by the receiver; and a setting unit configured to set a processing unit from which the extraction unit extracts a related document next, according to the processing structure received by the receiver.

Abstract: The present disclosure provides an intelligent interaction processing method and apparatus, a device and a computer storage medium. The method comprises: performing intention recognition for a preceding feedback item already returned to the user; continuing to return a subsequent feedback item to the user based on the intention of the preceding feedback item. According to the present disclosure, it is possible to guess the user's subsequent intention based on the preceding feedback item, and continue to return the desired subsequent feedback item to the user without the user's operations, so that the present disclosure is more intelligentized and richer and simplifies the user's operations.

Abstract: A computer-implemented method, system, and computer program product for generating a semantic model of data comprising a plurality of data items. Semantic concepts are identified for each data item in the plurality of data items. Data concepts are identified for each data item in the plurality of data items based on data characteristics of the data item and using a data ontology that defines the data concepts that are used to represent the data characteristics. The semantic model of the data is generated using the semantic concepts and the data concepts identified for the plurality of data items.

Abstract: An object recognition method is provided. The method includes obtaining speech information of a target object in a current speech environment and position information of the target object; extracting voiceprint feature from the speech information based on a trained voiceprint matching model, to obtain voiceprint feature information; obtaining a voice confidence value corresponding to the voiceprint feature information; and obtaining an object recognition result of the target object based on the voice confidence value, the position information, and the voiceprint feature information.

Abstract: A search unit (13) searches an ontology database (11) for a node set similar to a given word. A template creating unit (14) creates a template from common properties and nodes possessed by nodes included in the node set that has been searched for. An additional information determining unit (15) creates a triple by connecting the template to a node having the given word as a name, uses the created triple as display data, and when a triple is given, registers the given triple in the ontology database (11).